TY - JOUR ID - 42 TI - Development of an Efficient Hybrid Method for Motif Discovery in DNA Sequences JO - AUT Journal of Electrical Engineering JA - EEJ LA - en SN - 2588-2910 AU - Akbari, Reza AU - Zeighami, Vahid AU - Ziarati, Koorush AU - Akbari, Ismail AD - Corresponding Author, Reza Akbari is with the Department of Computer Engineering and Information Technology, Shiraz University of Technology, Shiraz, Iran, (email: akbari@sutech.ac.ir) AD - Vahid Zeighami is with the Department of Mathematics and Industrial Engineering, Ecole Polytechnique, de Montreal, Montreal, Quebec, Canada, (email: vahid.zeighami@polymtl.ca) AD - Koorush Ziarati is with the Department of Computer Science and Engineering, Shiraz University, Shiraz, Iran, (email: ziarati@shirazu.ac.ir) AD - Ismail Akbari is a graduated student from Department of Industrial Engineering, Iran University of Science of Science and Technology, Tehran, Iran, (email: ismail.akbari80@gmail.com) Y1 - 2012 PY - 2012 VL - 44 IS - 1 SP - 63 EP - 75 KW - Particle Swarm Optimization KW - Stochastic Local Search KW - Motif Discovery DO - 10.22060/eej.2012.42 N2 - This work presents a hybrid method for motif discovery in DNA sequences. The proposed method called SPSO-Lk, borrows the concept of Chebyshev polynomials and uses the stochastic local search to improve the performance of the basic PSO algorithm as a motif finder. The Chebyshev polynomial concept encourages us to use a linear combination of previously discovered velocities beyond that proposed by the basic PSO algorithm. Under this method, to balance between exploration and exploitation, at each iteration step, a local region is associated with each candidate particle, and a local exploration performed in this blob. The stochastic local search employs an intelligent repulsion/attraction mechanism to navigate a particle to explore this local region beyond that defined by the search algorithm to achieve a better solution. Over the successive iterations, the size of local region dynamically decreases. Also a non-linear dynamic inertia weight is introduced to further improve the performance of SPSO-Lk approach. The SPSO-Lk is tested on different sets of simulated and real nucleotide sequences to discover implanted DNA motifs. Experimental results show that the SPSO-Lk is effective, and provides competitive results in comparison with the performance of other algorithms investigated in this consideration.  UR - https://eej.aut.ac.ir/article_42.html L1 - https://eej.aut.ac.ir/article_42_13b72cc4f49e6c0d9459762baa124664.pdf ER -